Even after evidence-based treatment, Attention-Deficit/Hyperactivity Disorder (ADHD) is associated with poor long-term outcomes. These outcomes may be partly explained by difficulties in peer functioning, which are common among children with ADHD and which do not respond optimally to standard ADHD treatments. We examined whether peer rejection and lack of dyadic friendships experienced by children with ADHD after treatment contribute to long-term emotional and behavioral problems and global impairment, and whether having a reciprocal friend buffers the negative effects of peer rejection. Children with Combined type ADHD (N = 300) enrolled in the Multimodal Treatment Study of Children with ADHD (MTA) were followed for 8 years. Peer rejection and dyadic friendships were measured with sociometric assessments after the active treatment period (14 or 24 months after baseline; M ages 9.7 and 10.5 years, respectively). Outcomes included delinquency, depression, anxiety, substance use, and general impairment at 6 and 8 years after baseline (Mean ages 14.9 and 16.8 years, respectively). With inclusion of key covariates, including demographics, symptoms of ADHD, ODD, and CD, and level of the outcome variable at 24 months, peer rejection predicted cigarette smoking, delinquency, anxiety, and global impairment at 6 years and global impairment at 8 years after baseline. Having a reciprocal friend was not, however, uniquely predictive of any outcomes and did not reduce the negative effects of peer rejection. Evaluating and addressing peer rejection in treatment planning may be necessary to improve long-term outcomes in children with ADHD.